KR102591219B1 - Methods and test benches for performing test runs on test objects - Google Patents

Abstract

In order to ensure that the test object undergoes real environmental and/or ambient conditions, in particular thermal conditions, during the test execution on the test bench, at least one temperature is measured as a variable (MGi) measured during the test execution on the test bench (1) at the measurement point ( MSi), at least one test object component (PKi) of the test object (2) is divided into a plurality of segments (Si), and the thermal interaction of the at least one segment (Si) with the environment of the vehicle is simulated. The model 22 is simulated during the test run by a thermal simulation model 23, wherein the thermal simulation model 23 determines the segment heat flow (Si) supplied to or lost to at least one segment Si. ), and calculate the segment heat flow ( ) refers to the heat flow ( ) is regulated as a function of the temperature measured on the test bench 1 on at least one segment Si by a plurality of heat flow actuators 15 _j such that it undergoes

Description

Methods and test benches for performing test runs on test objects

The present invention relates to a test bench and a method for carrying out test execution on a test bench, wherein a test object in the form of a car or a part of a car is set up and is actually operated on the test bench, and a simulation unit with a simulation model is used to perform the tests. Simulate execution.

A test, trial or experiment refers, in general terms, to whether one (or several) technical component(s) (mechanical setup, hardware or software) being tested functions in the context of specific framework conditions and/or determines whether a specific This is an activity used to determine whether a feature exists. Therefore, the test object is the technical system that is to be tested. The test object may be an overall system (e.g. a vehicle) or a part of an overall system (e.g. a combustion engine, a drive system, an exhaust gas system or an exhaust after-treatment system of a vehicle).

Some tests are used to examine the actual, often transient, process that some part of the test undergoes. These tests are often conducted in a reproduced environment, that is, a simulated environment. Dedicated test benches are often used for these activities, such as engine test benches, drivetrain test benches or roller test benches. These test benches systematically subject the test object to specified environmental conditions, which means that the test object is tested under private environmental conditions, thereby enabling replay of the test process. However, test benches are also used for other purposes, such as verifying processes by new, unknown or very time-consuming and expensive processes in environments that cannot be physically reproduced against the test object. Although a test environment is always an imperfect representation of the real environment, any results obtained from testing should always be analyzed in light of the quality of the test environment, i.e., the quality of the test bench and the simulated environment.

For example, a given object on a test bench often simulates real or fictitious (i.e. virtual) test execution of an automobile. Hereinafter, these runs will be referred to as virtual test runs. For example, a combustion engine on an engine test bench or a drive train on a drive train test bench can be varied over time through an appropriate interface (the interface between the real and the test bench available for simulation) and the actual test drive. While the overall system car/driver/environment, for example, the configuration of a real car driving on the Großglockner High Alpine Road, interfaces with variables experienced by the test subject. Likewise, it may be interesting to allow the test object on the test bench to vary over time, with variables that may arise from arbitrary courses, or courses that are fictitious and do not necessarily represent actual routes. These virtual test runs may be generated by different processes. For example, they may be measured during actual test runs, or they may be partially predefined and/or standardized (eg, standardized consumption cycles). However, such execution can be performed in real time or near real time (i.e. online) by means of a virtual environment (so-called X-in-the-loop test execution, where " ) can also be calculated as sufficiently good quality. Various measurements can be taken on the test bench during the execution of a virtual test run. For this purpose, the test object to be inspected (combustion engine, drive train, battery, vehicle, etc. or their components) is tested by a load unit (actuator), for example a dynamometer (mechanical actuator) or a battery tester (electrical actuator). By loading the interface, the test object is subjected to mechanical or electrical loads resulting from virtual test execution. The execution of such tests on a test bench is, in particular, the execution of development or testing tasks on the test object on the test bench, establishing an overall system (e.g., a physically complete automobile) into which the test object is normally integrated as part of the overall system. This makes it possible without first requiring complete testing, which would have to be done by actual test runs on real cars. Moreover, these tests on the test bench have the advantage of excellent reproducibility and therefore excellent compatibility of results.

However, due to system limitations, the test bench is only able to reproduce the conditions of real test execution with certain limitations as well as assured precision. Additionally, it is not always desirable or required for test subjects to be subjected to these exact conditions in all cases. In some cases, test subjects may simply think and experience hypothetical conditions. According to the prior art, current work on test benches, for example, focuses substantially on a high level of correspondence of mechanical and electrical power flows between virtual and real test runs. Despite the good correlation of a particularly measured variable over time, taken on a test bench, there are often significant differences for some other variables measured over time. For example, despite the excellent quality of consistency of mechanical power flow (e.g. speed and torque) and the use of the same measurement technique to take dissipation measurements during virtual test runs on a drive train test bench and a roller test bench. And when comparing the results for actual test runs with a physical car, it was found that emission measurements (CO, NOx,...) do not produce the same results, especially in the partial load range. The reason was found to be differences in the field of temperature and heat flux density for different thermal and thermodynamic conditions, i.e. neuralgia automotive parts. The differences are due, among other reasons, to the different media flows (e.g., water, air, oil, etc.) experienced by the test object. A real vehicle contains components and/or structural parts that may be exposed to various heat loads. A related example is a combustion engine with a turbocharger and exhaust gas system. Components such as catalytic converters and particulate filters may also be included in the exhaust gas system. The flow of heat energy over the surfaces of the mentioned parts and sub-components occurs in both cases (real test execution or virtual system execution). Electrical energy storage in hydrogen cars is another example. Additionally, these components thermally interact with the environment/surroundings and are themselves affected by influences from the environment/surroundings (“conditions”). Accordingly, the engine block or exhaust gas system of a vehicle will engage in heat exchange with its environment in various ways, depending on the conditions and/or ambient conditions. As a result, there are therefore different transient part-temperature fields and heat flux densities (thermal energy flows) inside the part and on its surface. For example, in winter-like environmental conditions (cold road conditions, cold ambient air), a combustion engine will radiate a greater amount of heat than in summer-like environmental conditions (hot road conditions, hot ambient air). Heat transfer (heat flux density) between the test object and the environment occurs based on the physical mechanisms of thermal gradients, heat flow (convection), and heat radiation.

During a real test run with a real car along a real route, the heat transfer process is virtually on the surface of the test object and/or on the parts of the test object. Variable environmental and/or ambient conditions are characterized by barometric pressure, humidity or temperature in nature due to, for example, water spraying on the test subject. However, various environmental conditions exist on the test bench in practice, which is one reason why results from virtual test execution on the test bench may deviate from actual test execution. The generation (simulation and/or emulation) of heat transfer processes on a test bench given arbitrary conditions has hitherto been of little interest and/or has not attracted sufficient attention so far.

Cooling-air blowers, often used on test benches to create airflow over the test object, as well as test bench conditioning (e.g., controlling the temperature within the test bench space) are typically used to accurately emulate real-world environmental conditions on a test bench. It is insufficient. Cooling-air blowers are commonly used to reproduce the effect of engine cooling on headwind speed. This is why cooling-air blowers are often improperly dimensioned and/or they do not provide the required degree of freedom. For example, frequently, the speed of a cooling-air blower is controlled only as a function of travel speed. By providing air conditioning within the test bench space, the air temperature and humidity within the environment of the test bench can be controlled.

The use of conditioning equipment for media on the test bench, such as intake air, coolant, oil and charge air, is more known. The above equipment is mostly used for component test benches (engine, drive train, battery test bench, etc.). By means of this equipment, the respective temperature of the medium is influenced and/or controlled. Furthermore, conditioning equipment for intake air can affect air humidity and pressure.

Cooling-air blowers and testing room air conditioning, as well as media conditioning, receive setpoint settings such as temperature, humidity and pressure from the test bench automation system. Settings are set once for each variable (e.g., temperature) over time, and there is no interaction (in the sense of an This may also be an expected future environment. In addition, there is the problem of finding settings, that is, setting settings that reflect the virtual environment conditions of the test object in a way that is close to reality.

To better emulate thermal conditions, test bench arrangements with thermally encapsulated parts under test (eg, an engine on an engine test bench) are also known in the art. Devices of this kind are described, for example, in Kramer S., et al., “Verlagerung von Rollentests auf den Motorenprufstand,” MTZ-Motortechnische Zeitschrift, 2015, 76 (3), pp. It can come from 36-41. The vehicle body is positioned on an engine test bench in that the engine is placed within the enclosed engine encapsulation and the exhaust gas system is positioned within the enclosed lower encapsulation to simulate thermal conditions within the engine compartment and/or underbody. It is simulated. An enclosure is provided in an isolated housing (engine encapsulation, bottom encapsulation) that is fitted with the blower. The temperature within the engine and lower encapsulation is controlled by the blower. Therefore, the solution as introduced here allows comparison of results from measurements of dissipation on an engine test bench with results from a roller test bench. However, this is insufficient for a realistic reproduction of the environmental conditions, since, on the one hand, the global temperature is regulated in the engine and lower encapsulation, and on the other hand, problems associated with setting the correct setpoints persist. As a result, various components under test, such as engine blocks, turbochargers, coolers, exhaust gas systems, etc., and/or parts thereof, have temperature distributions on their surfaces that do not match the actual or desired temperature distribution. Nevertheless, the temperature of these components not only has a decisive influence on the heat transfer process in the form of thermal energy flow (“heat flow”), but also influences the dissipation behavior of the engine (e.g. NOx, CO, etc.). , which results in unwanted inconsistencies between actual and virtual test executions. This means that the described method does not solve the problem of reproducing and/or predicting the thermal behavior of the test object during actual test execution.

Therefore, the patent DE 10 2013 213 863 B3 already describes a cooling system for a component such as a combustion engine, which is capable of regulating the temperature of the component, so that the component is ventilated by a blower matrix consisting of a plurality of individual blowers. The cooling system within it can regulate various temperature regions (temperature fields) on the component. The target temperature for individual points is preset as a setpoint, which is known from the very beginning (meaning the earliest possible start of the test run), for example, a blower matrix to check the heat intensity of a part or part of a part. It is the shape of the time curve that is controlled by the controller unit through. In contrast to conventional test bench devices, this is an improvement especially with regard to thermal conditions for test bench experiments, which can often be sufficient. However, component temperature as a target variable for controlling conditions on the test bench ignores the heat transfer process in the form of thermal energy flow of the actual test object within the various test environments. Effects such as convection, thermal radiation, etc., which play an important role in the components under test (the physical car), are therefore omitted from consideration on the test bench. DE 10 2013 213 863 Providing a temperature field of the surface of the test object and/or of the test object itself, as described in B3, neglects the heat transfer process and thus requires a realistic test bench in the form of virtual test execution on the test bench. Often insufficient for experimentation.

Accordingly, patent DE 10 2013 213 863 B3 is based on the very restrictive assumption that the target temperatures for the selected measuring points are known as a function of time (i.e. these temperatures can be set in advance as command variables of the control). The values must be predefined; however, such arbitrary decisions may not enable the creation of realistic environmental conditions, or the values would have to be established in advance over the course of an expensive and complex actual test run. Patent DE 10 2013 213 863 B3 does not solve the problem regarding the determination of the set point.

Therefore, it is an object of the present invention to provide a method and associated test bench for executing a test run, for example in the form of a thermal A test object is placed on a test bench that causes the test object to undergo conditions that result based on the conditions and/or laws of thermal heat transfer of the virtual environment.

This object is achieved according to the invention by the method specified in the introduction and in the same way by the test bench mentioned in the introduction, wherein at least one temperature is measured as a variable measured at a measuring point on the test bench during the test run, At least one test object component of the test object is divided into a plurality of segments, and the thermal interaction of the at least one segment with the environment of the vehicle is simulated during test execution by a thermal simulation model of the simulation model, calculates the segment heat flow supplied to, or dissipated from, at least one segment, such segment heat flow being transferred to the test bench on the at least one segment by means of a plurality of heat flow actuators that subject the test object to the heat flow; is adjusted as a function of the measured temperature.

Because of the present invention, it is therefore possible to subject the test object to certain thermal environmental conditions, creating conditions on the test bench that approximate reality. These thermal environmental conditions interact with the test object in the form of heat transfer processes, and they can approximate reality (i.e. they may in fact occur in the later reality of the test object as part of a car that is actually used - e.g. , a car traveling through Death Valley), or they are fictional (i.e. considered, but still correspond to real physical conditions, e.g. a car traveling through Death Valley at an external temperature of 60 C). ) can be. Aside from thermal loads, test objects are frequently additionally stressed on the test bench, with mechanical and electrical loads and/or through mass and information flow (e.g. CAN communications).

Based on the thermal simulation model, it is possible to reproduce the thermal interaction of the test object and/or segments of the test object adjusted to the simulated environment in any desired way within specific technological boundaries. It can simulate heat transfer processes in a test object that vary over space and time and correspond to real-world conditions. Using heat flow actuators, it is possible to regulate these heat transfer processes on the test bench to the test object, essentially under real or fictitious conditions (but still physically identical to reality, e.g. During a test run (moving from the thermal conditions of Death Valley to the thermal conditions of Antarctica within two hours), partial components of the overall system (e.g., a car) are subjected to identical or sufficiently similar thermal conditions.

Preferably, the quality of the simulation can be further improved if the simulation model additionally includes one or more of the following models: car model, driver model, road or route model, wheel model, environment model. Moreover, additional partial models are suitable to increase flexibility and allow the most variable influences to be taken into account during test execution.

Preferably, at least one additional measured variable is detected and processed in the simulation model. It is likewise further advantageous if at least one additional measured variable of the test object's environment under test is detected and processed in the simulation model.

The invention, which is the subject of the present specification, will be described in greater detail below with reference to Figures 1 to 6, which are schematic and exemplary and are not limited to the preferred embodiments described according to the invention. It appears as follows:
Figure 1 shows a roller test bench for an automobile according to the prior art.
Figure 2 shows a roller test bench according to the invention.
Figure 3 shows an engine test bench according to the invention.
Figure 4 shows control of heat flow over a segment of a test object according to the invention.
Figure 5 shows an embodiment of a simulation unit.
Figure 6 shows the information flow in the execution of the method according to the invention for operating a test bench.

Figure 1 shows a conventional test bench (1) for a test object (2). In the illustrated embodiment, the test object 2 is a car and the test bench 1 is a roller test bench. The test object 2 may be any partial system of the automobile, such as the drive train, combustion engine, drive battery (power pack), turbocharger, ignition converter, etc., and the test bench 1 may be any matching test bench, It should be understood that this could be, for example, a drive train test bench, an engine test bench, a power pack test bench, a turbocharger test bench, a fuel converter test bench, etc.

Test bench automation is provided in the test bench 1, which takes the form of a test bench automation unit 3 that controls virtual test executions, which are executed on the test bench (=test execution), so that the test execution Activate all devices of the test bench (1) essential for this (i.e. especially actuators) according to the requirements. The test bench automation unit (3) can in particular operate the test object (2). If the test object 2 is a car, for example, a known driver robot can be placed inside the car, which executes control commands such as gear shifting, acceleration, etc. by the test bench automation unit 3. Alternatively or additionally, the test bench automation unit 3 may directly, for example, via the control unit under test (e.g. automotive control unit (ECU), transmission control unit (TCU), hybrid control unit, battery management system, etc.). You can operate the test object (2) with . If a combustion engine is the test object 2 , the test bench automation unit 3 can, for example, operate the throttle valve position α (see FIG. 3 ) or fuel injection.

The load, when viewed mechanically (mechanical power flow between the test object and the environment), is applied to the test object (2) by a load machine (usually an actuator) (5). In the case of a roller test bench, the (mechanical) load machine 5 is the input and/or output of the test bench roller, as indicated in FIG. 1 . In the case of a combustion engine or drive train as test object 2, the mechanical load machine 5 will be, for example, a dynamometer or an electrical dynamometer connected to the combustion engine or drive train. If a battery is the test object 2, the load machine 5 will be electrical, for example in the form of an electrical battery tester. Suitable load machines for various test objects 2 are well known in the art, which is why no further discussion needs to be presented.

The load machine (5) is largely controlled by the actuator controller (4), which in turn operates the test bench automation unit (3) to regulate the instantaneous load moment (M) or especially the instantaneous speed (n) of the test object (2). It receives the set point from . The test bench (1) is a torque measuring device that establishes corresponding actual values of the load moment (M) and speed (n) of the test object (2) and makes these values available to the test bench automation unit (3). (6) and/or a speed measuring device (7) are also typically provided. Other or additional measured variables, such as measured electrical currents or electrical voltages, can be used for other test objects 2 and/or test bench types, and these are fed into the test bench automation unit 3 .

Furthermore, emissions are measured on the test bench 1 during test execution, for example by means of an exhaust gas measurement system 14 . Naturally, depending on the test object 2 , other or additional measurements may be taken, such as certain measurements essential for the development of the test object, such as measurements of consumption, electrical energy flow, etc. The basic purpose of test execution is to detect and analyze at least one output variable of the test object 2, such as dissipation, consumption, power, etc., which are derived on the basis of the development for the test object 2. . This effort ensures that the test object 2 behaves essentially the same on the test bench 1 and when it is desirable to integrate it into the physical vehicle.

At least one adjustment unit 16 is often provided on the test bench 1 for adjusting the test object 2 and/or the test object environment of the test object 2 . In particular, in this way it is possible to cause the test object 2 to undergo a specific (e.g. desired) heat transfer, which heat transfer is variable in time-light space, and the test object 2 on the test bench 1 Exchange heat transfer with the testing environment of the test object. Heat transfer may be coupled to a specific mass transfer, such as heat transfer with an air flow or another mass flow. Therefore, heat transfer includes this mass transfer which is equivalent to heat transfer. In a testing environment, test bench air conditioning means for controlling environmental temperature, humidity, etc. are often provided as a control unit 16. Furthermore, the regulating unit 16 may also include a blower 8 to simulate, for example, a headwind. Additionally, the blower 8 can be installed separately and provided separately from the regulating unit 16 on the test bench 1 . A blower 8 of this kind allows the test object 2 to undergo a specific (e.g. desired) heat transfer process that varies in time and space and allows this test object 2 to exchange heat transfer with the testing environment. contribute to doing It makes perfect sense that different adjustment units 16 may frequently be used for different types of test benches. In the illustrated embodiment, the conditioning unit 16 comprises a blower 8 which causes the test object 2 to experience a specific air flow field 9 .

Furthermore, in a manner known in the art, the adjustment unit 16 of the test bench 1 for adjusting the test object 2 can be used for media adjustment, for example intake air conditioning, charge air conditioning, oil conditioning or coolant conditioning. Additional units may be included. The former is not shown in Figure 1 to increase brevity of the drawing. Additionally, these actuators serve to subject the test object 2 to specific, often desirable, heat transfer processes that vary over space and time.

The conditioning unit 16 applicable to the blower 8 and/or the media conditioning unit may be configured to provide a specific Setpoints (temperature, humidity, mass flow, etc.) are typically received. As outlined in the introduction, the blower 8 and/or the conditioning unit 16 with a typical media conditioning unit simulate the desired (i.e. realistic approximation) heat transfer process for the test object 2 or the component under test. cannot be done or simulated inadequately (in terms of the testing mission to be accomplished).

.According to the invention, the thermal conditions of the test object 2 are preset during the test run or in a way that simulates the desired conditions, in particular those that match reality (in the sense of a “tracing run” of an actual test run on a test bench). , it is provided that the thermal conditions of the test object 2 are reproduced in accordance with these requirements, in order to carry out the test run on the test bench 1. This is based on Figure 2 and an example of a roller test bench as a test bench (1) and a car as a test object (2) and with reference to Figure 3 an example of an engine test bench as a test bench (1) and a test object (2) ), the combustion engine will be described below. From Figure 1, some parts of the test bench 1 have been omitted to improve the brevity of the drawing.

The test object 2 includes a plurality of test object components PKi, i = 1, ..., m, where the test object component PKi is the combustion engine 10, the exhaust gas system 11, or It may be a complete assembly of the test object (2), such as an exhaust gas after-treatment unit (12, 13), such as a catalyst converter or a particle filter within the exhaust gas system (11). However, the component under test PKi may be a part of the under test 2 or a part of an assembly of the under test 2, such as an exhaust pipe section of the exhaust gas system 11. However, for example, when the test object 2 is an electric rechargeable battery, the entire test object 2 may represent the component under test PKi (i=1). The invention enables the presence of at least one such component under test (PKi). A component under test in the sense of the invention is, as described below, in particular a part of the test object 2, which undergoes thermal interactions (heat transfer, heat flux density) that vary over space and time. This test object (2) exchanges thermal interaction with the test object environment. Therefore, these parts are particularly suitable as components under test (PKi), whose behavior or characteristics are a function of thermal load. In this way, it is possible to influence certain characteristics of the test object 2. For example, the characteristic “NOx emissions” is, among other things, a function of the thermal load of the component under test “catalytic converter”.

In the number n of measuring points MSi (i = 1, ..., n), according to the invention, at least one measuring point MS1 is required, and measuring units MEi, i = 1, ... , n) are prepared, whereby the measurement variable (MGi, i = 1, ..., n) of the test object 2 is measured. Different measurement units (MEi) may be provided at one measurement point (MSi) for measuring different measurement variables (MGi). At least one measured variable (MGi) therein is temperature or a measured variable, and the temperature can be calculated or estimated based on this. Accordingly, the at least one measurement unit MEi is, for example, a simple temperature sensor that serves to measure the temperature of the test object 2 at the measurement point MSi. The measuring unit MEi for detecting temperature can detect, for example, the temperature of a medium such as exhaust gas or fluid temperature, the assembly temperature, and/or the component temperature or surface temperature. In principle, it is also possible to measure the complex three-dimensional temperature field of the component under test PKi or a part of the component under test of the test object 2 by means of a thermal image camera as a measurement unit MEi or a multifms method.

Applying appropriate mathematical/physical methods, the test object or the part under test (PKi) or a part thereof (e.g. the surface of the part under test (PKi)) is determined based on any temperature measurements taken on the test object (2). The entire temperature field (spatial temperature distribution) can be derived. Methods of this type can estimate temperatures and/or spatial temperature curves between measurement points MSi using known interpolation methods, for example via spline functions, or finite element methods.

Using a measuring unit (MEi) on the test object (2), it is also possible to measure the medium flow, such as the flow of exhaust gases through the exhaust gas system (11) or the intake air flow. Other possible measurements of the medium pressure, such as exhaust gas pressure, can also be taken at various points.

Likewise, with the measurement unit MEi on the test bench 1, it is possible to additionally measure measurement variables MGi in the test object environment of the test object 2, preferably in the immediate vicinity of the test object 2. The measured variable (MGi) of the test environment may be, for example, atmospheric pressure, environmental temperature, or humidity. To improve the brevity of the figures, not all measurement points (MSi), measurement units (MEi) and measurement variables (MGi) are shown in Figures 2 and 3.

In order to simulate the heat transfer process in a desired or predefined manner for the component under test (PKi) on the test bench (1), at least one heat flow actuator (15 _j , j = 1, ..., k) is provided. provided. The desired heat transfer process is generated on the component under test PKi by at least one heat flow actuator 15j, in particular the heat flow ( ) or heat flux density ( ) is in the form of For simplicity, only class will be used below. heat flow ( ) is the heat flux density ( ), and the two variables can be used in an equivalent manner. Therefore, heat flow ( ) will be used hereinafter, in an equivalent manner the heat flux density ( ) or heat flow ( ), it should also be understood to include any other variables that are equivalent.

Heat flow actuator 15 _j may represent a heat sink, a heat source, or both. Various devices may transfer heat (in any direction) or, in particular, the component under test (PKi) may cause heat flow ( ) can be considered as a heat flow actuator (15 _j ) that allows it to experience. Water or air heat exchangers, fluid flow devices (e.g., blowers, venturi flow devices), Peltier elements, spray nozzles for spraying fluids such as water, etc. are considered devices. This therefore means that, as a heat flow actuator 15 _j as shown in FIGS. 2 and 3 , it is in principle possible to use even a conventional adjustment unit 16 for test bench adjustment. In the same way, heat flow actuators 15 _j as shown in FIG. 3 with heat flow actuators 151 , 152 and 153 , for example for intake air conditioning, charge air conditioning, oil conditioning or combustion engine 10 It is possible to use the media control unit of the blower (8) of the test bench (1) or the control unit (16), such as coolant control of . Media conditioning units of this type typically consist of a heat exchanger for each media. Accordingly, this means that the test object 2 and/or the component under test PKi is subjected to a specific, preferably predetermined, heat transfer that varies over space and time, which heat transfer causes the test object 2 ) is the heat flow actuator (15j) and the heat flow ( ), it is exchanged with the test target environment.

The correct configuration of the heat flow actuator 15 _j is incidental to the purpose of the present invention. The specified requirements for the heat flow actuator 15 _j are: heat flow ( ), and/or heat flow ( ) can only be created. This means that any heat flow actuator 15 _j can supply heat to, and/or dissipate heat from, the test object 2 .

Using the measuring unit MEi, when the heat flow actuator 15 _j is the blower 8 (e.g. see FIG. 2 with the measuring unit MEn) the blower speed or the flow rate of air or the heat flow actuator ( When 15 _j ) is a heat exchanger, it is possible to detect a measured variable of the heat flow actuator 15 _j , such as the fluid flow of the heat exchanger fluid (air, water, etc.).

At least one heat flow controller 17 controls the desired heat flow ( ) deals with the operation of the heat flow actuator (15 _j ) to regulate Implementation of the heat flow controller 17 may be as a separate unit on the test bench 1 (as seen in Figure 3), and/or integrated within the heat flow actuator 15 _j , and/or in test bench automation. This can be achieved as part of unit 3 (as can be seen in Figure 2).

Therefore, controlling the heat flow actuator 15 _j by the heat flow controller(s) 17 is a multi-variable control, which involves controlling at least one measured variable MGi, in particular the specific heat flow ( ), the temperature at the measurement point (MSi) of the test object (2) is processed. If a measuring unit MEi for detecting the actual variable is needed to control the heat flow actuator 15 _j , a corresponding measuring unit MEi must be provided. As an alternative, the essential actual variable can be calculated based on another measured variable (MGi). Any suitable control law may be implemented in the heat flow controller 17, and the specific implementation of the control law is incidental to the purposes of the present invention.

The measuring units MEi supply their measured variables MGi to the heat flow controller 17, which processes the corresponding measured variables MGi, which, if necessary, provide these variables to the test bench automation unit ( 3) and is also supplied to the simulation unit (20).

The effect of the heat flow actuator 15 _j on the individual component under test (PKi) to be controlled is usually coupled. This means that the heat flow actuator 15 _j operates simultaneously on a plurality of components under test PKi, and that the component under test PKi is influenced by a plurality of heat flow actuators 15 _j at the same time. do. Therefore, in order to control the heat flow actuators 15 _j , it is desirable to decouple the individual heat flow actuators 15 _j . Many references were found in the literature describing relevant known methods in the technical field (e.g. JK Hedrick, A. Girard, "Control of Nonlinear Dynamic Systems: Theory and Applications,"2005; here: particularly Chapter 8 and S. Skogestad) , I. Postlethwaite "Multivariable Feedback Control - Analysis and Design, 2nd Edition, 2001; here: particularly Chapters 9, 10 and 3.4.1), which is why it is not discussed in more detail.

As long as there is no coupling between the different components under test (PKi) and each assigned heat flow actuator (15 _j ), decentralized control can be achieved without decoupling with a stand alone heat flow controller (17).

In addition, the heat flow actuator 15 _j is a spatiotemporally variable heat flux field ( ) or a heat flux density field similarly acting on the component under test (PKi) ( ) is created in the test object (2). When the heat flow actuator 15 _j is decoupled, the heat flux field ( ) or similarly the heat flux density field ( ) gives birth to

The component under test (PKi) is preferably divided into finite segments (Si) with i = 1, ..., s. The division into segments Si can be carried out with a granularity suitable to the requirements or application. The segment Si may be a complete component under test PKi, such as the exhaust gas system 11 or the exhaust gas aftertreatment units 12, 13 of the exhaust gas system 11. Still, the segment Si may be divided into finger segments in the same manner, for example, the component under test PKi may be divided into a plurality of segments Si. For example, the exhaust gas system 11 can be divided into 10 segments Si. However, in principle, the entire test object 2, for example a battery, could also be a segment Si. The specific division of segments Si is incidental in the context of the present invention. However, it is important to note that the number of i = 1, ..., k of heat flow actuators 15 _j need not match the number of i = 1, ..., s of segments Si. In fact, typically, there will be no match. The heat _flux field ( ) Because, therefore, segment-heat flow ( ), which means the heat flow from the environment under test to each segment (Si), or from each segment (Si) within the environment under test. Figure 4 is a schematic diagram in this embodiment, and the component under test (PKi) was divided into six segments (Si).

The heat flow actuator 15 _j is configured to flow heat to and/or from the component under test PKi ( ) is created. As previously described, the measuring unit MEi is provided at a certain measuring point MSi of the test object 2 and, if necessary, is provided in the environment of the test object 2 so that at least the temperature is adjusted to the test object 2. is measured for. Using the measurement unit (MEi), it is possible to detect a measured variable (MGi) of the test object (2) and/or the component under test (PKi) and to test the test object (2), such as air pressure or humidity in the testing space. It is also possible to detect a measured variable (MGi) of the target environment or a measured variable (MGi) of the heat flow actuator 15 _j , such as flow rate. The measured variable MGi detected by the measuring unit MEi is fed to the heat flow controller 17, which controls the desired segment heat flow ( ) (setpoint setting), the manipulated variables for the heat flow actuator 15 _j are now calculated according to the implemented control law. Therefore, in a targeted manner, the segment heat flow ( ) can be adjusted. Naturally, segment heat flow ( ), the heat flow actuator 15 _j is placed on the test bench 1 in a way that allows adjustment of ).

Therefore, the test object (2) on the test bench (1) has a heat flux field ( ) is described, a flexible, open and scalable generalized IO system (including sensors, actuators and controllers) capable of undergoing Provides an appropriate level of corresponding quality and dynamics.

According to the invention, the segment heat flow ( ) to generate the heat flux field ( ) setpoint settings are provided. Based on at least one suitable simulation model 22 with “real-time capabilities”, this simulation unit 20 is able to simulate a heat flux field varying over space and time ( Segment heat flow regulated through ) ( ) creates a set point in the form of

Now, it is possible to perform virtual trial runs (test runs) where a real test object (2) is integrated into the overall virtual world of the car and simulated in the environment of the virtual world (X-in-the-loop simulation). This means, for example, that the simulation model 22 moves a virtual car through a virtual world. Additionally, the simulation unit 20 can be run on a test bench automation unit 3 . The simulation for virtual test execution on the test bench 1 is preferably real-time, i.e. using the heat flow actuator 15 _j to determine the required heat flux field ( ), a current setpoint value is calculated for each time increment, for example in the range of milliseconds to minutes.

The simulation model 22 includes at least one thermal simulation model 23, as shown in FIGS. 4 and 5, wherein the test object 2 or the component under test PKi is contained within a physical vehicle. If the vehicle moves along a predefined route, it simulates how the test object 2 and/or the component under test PKi will thermally interact with the environment. This environment of the component under test (PKi) consists of vehicle components that are not available on the test bench (e.g. adjacent vehicle assemblies or parts) and the vehicle's environment (e.g. airflow, road surface,...). Therefore, the thermal simulation model 23 simulates automotive components that are not physically present on the test bench 1 (and, if necessary, components included on the test bench) and the environment (airflow, application surfaces, etc., e.g. It specifically simulates the thermal behavior of the hood shape and lower model). This thermal interaction manifests itself as a heat flow, which is simulated on the test bench 1 with the help of a heat flow actuator 15 .

Additionally, a car model 24, a driver model 25, a road or route model 26, a wheel model 27, etc. may be executed in the simulation model 22 shown in an exemplary manner in FIG. 5 . You can also run an environmental model that simulates the car's environment. The different partial models of the simulation model 22 work together to execute the test run, taking into account the thermal interaction of the test object 2 and the vehicle's environment. The simulation model 22 can be used to simulate different types of driver behavior (conservative, aggressive, etc.), road conditions (e.g., rain, ice, different road pavements, etc.) or other influences such as tires. Therefore, the heat flux field ( ) can occur due to simulated driving situations that account for specific environmental conditions. For example, a sporty driver will cut curves and thus drive over ice floes or puddles (such as water splashes) on the road, while a conservative driver will follow the lines of the road and drive over ice floes or puddles. It can be assumed that will be avoided. This has a direct effect on the heat transfer process to the test object (2). On the right side of the test bench 1, it is possible to provide actual control elements for the car, such as steering wheel, gas pedal, brake pedal, gear shift, etc., which can be used to actively intervene in the test run. Preferably, the simulation is run in real time and with the required time resolution.

Instead of various partial models (car model (24), driver model (25), road or route model (26), wheel model (27), underhood and underbody models as part of a thermal simulation model (23), etc. Test execution can be provided in a variety of ways, for example in the form of conventional time-based or path-based speed settings. The specific test execution is determined on the basis of a partial model or a time-based or path-based setup, in which the thermal interaction of the test object 2 with the environment is simulated by a thermal simulation model 23.

The simulation unit 20 further comprises an interface 21 (FIG. 5), whereby the measured variables MGi required for the simulation model 22 as well as the real variables of the test object 2 (e.g. one or more real variables) are used. Speed (n _ist,x ) or load machine (5), one or more Selje torque (M _ist,z ) can be supplied, and the simulation model (22) is the test object (2) (e.g., throttle position α _soll ) and /or test bench 1 (e.g., setpoint torque M _soll and/or setpoint speed n _soll of load machine 5 or a plurality of setpoint torques and/or setpoint speeds for a plurality of load machines) and/or output calculated setpoint values, in particular for controlling the test execution of the heat flow actuator 15 _j . If necessary, the interface 21 provides the necessary signal processing mechanisms, such as filters for the measured variable MGi. By providing the measurement variable MGi to the simulation unit 20, the “simulation loop” is closed and the test object 2 is effectively integrated “in the loop” of the virtual-real world.

The thermal simulation model 23, which reproduces the thermal interaction of the component under test (PKi) with the environment, can be formed, for example, in the form of a physical model, an experimental model or a trained model (neural network, linear model network, etc.), It can be designed in this way. Furthermore, the thermal simulation model 23 is a component under test (PKi) that is physically present on the test bench 1 and is inspected to reconstruct unmeasured or unmeasurable variables (e.g. temperature) (e.g. by a control observer). ) can be reproduced. With each preset time increment, the thermal simulation model 23 determines the segment heat flow ( ) Determine the set point for . For this purpose, the thermal simulation model 23 processes at least one temperature (or another equivalent physical variable) that was measured by the measurement unit MEi at the relevant measurement point MSi. It is understood that the thermal simulation model 23 may also process measured variables MGi, such as mass or volume flow, barometric pressure, ambient temperature, etc. The required measurement variable MGi will depend on each run of the thermal simulation model 23 and, if necessary, on each run of the other models of the simulation model 22. The measured variables (MGi) essential to the thermal simulation model 23 may not be obtained on the basis of direct measurements, but may be estimated on the basis of other measured variables (MGi), for example through appropriate observers or by calculating observers. there is. Using the example of exhaust gas system 11, based on measurements of the input and output temperatures of the exhaust gases entering and exiting exhaust gas system 11 and measurements of the mass of exhaust gas passing through exhaust gas system 11 , surface temperatures at different locations on the exhaust gas system 11 can be calculated.

The thermal simulation model 23 may further process variables of the test run, such as those obtained from other models in the simulation model 22 or from a set point speed, such as vehicle speed. Preferably, the test run will be preset environmental conditions, such as air temperature, humidity, etc., which may be included in the simulation model 23. However, events that can be included in the thermal simulation model 23, such as lightning, overtime in a car or driving through a puddle, can also be preset.

Via the interface 21, the segment heat flow (e.g. determined numerically or model-based) ) are fed to the heat flow controller 17, which adjusts these set points by means of a plurality of heat flow actuators 15 _j , each at preset time increments, to a particular level of control quality. , at least one heat flow actuator 15 _j is provided in at least one segment Si, preferably in all segments Si. Its quality depends, among other factors, on the specific implementation of the heat flow actuator 15 _j . For this purpose, the manipulated variable for j available to the heat flow actuator 15 _j is the segment heat flow in the heat flow controller 17 according to the implemented control law ( ) is calculated based on the set point of the corresponding heat flow ( ) and/or heat flux field ( ) is preset for the heat flow actuator 15i, which generates

The following example seeks to illustrate the method according to the invention. The test object (2) is integrated into a real car and moves across a real testing terrain in the context of a real trial run. This refers to the specific actual segment heat flow ( ). The target is now a set point from an appropriate thermal simulation model (23), a virtual trial run on the test bench (1), i.e. these real segment heat flows ( ), which occurs during the actual trial run. According to the laws of physics, the segment heat flow ( ) significantly depends on the resulting temperature field of the test object 2, which is detected via the measuring point MSi, for example by heat conduction, convection, heat radiation. For this purpose, the temperature of the test object 2 is measured at every time increment on the measurement point MSi, based on a thermal simulation model 23, and the segment heat flow ( ) is calculated and adjusted on the test bench 1 by the heat flow controller 17 and the heat flow actuator 15 _j .

It should also be noted that in general there need not be a 1:1 correspondence between measurement points (MSi) and segments (Si). For example, for an individual segment, it is possible to measure the temperature several times, whereas, on the other hand, for different segments (Si), not all temperature measurements are required. For these segments, the temperature field is only estimated in these examples.

Segment heat flow ( The information flow for controlling ) is shown once more in generalized form in Figure 6. A plurality of measurement variables MGi are detected at specific measurement points MSi of the test object 2 using measurement units MEi. The temperature (or equivalent physical variable) is detected at least at one measurement point MSi of the test object 2 therein. It is also possible to further measure mass or volume flows as described above, as well as other measured variables (MGi), such as environmental variables (ambient temperature, humidity, barometric pressure, etc.). Via the interface 21 the measured variables MGi are supplied to the thermal simulation model 23 of the simulation model 22 in the simulation unit 20 and can also be supplied to additional models of the simulation model 22 . Based on the measured variables MGi, the thermal simulation model 23 determines the segment heat flow on the segment Si ( ) Determine the set point. Segment heat flow ( These set points of ) are fed to the heat flow controller 17 for regulation by the heat flow actuator 15 _j . The heat flow actuator 15 _j is the necessary associated heat flow ( ), which acts on the component under test (PKi) or segment (Si) to be adjusted.

Segment heat flow ( ) changes rapidly with each successive time increment, for example when driving through a puddle of water, a large amount of water from the puddle evaporates from the hot muffler of the exhaust gas system 11 and the heat flow actuator 15 _j ) due to limited dynamic properties, segment heat flow ( ) There may be cases where such rapid changes cannot be controlled. In these cases, at least the segment heat flow ( ) is adjusted to the integral mean over a suitably chosen time window, such as 1 minute, so that the heat exchanged integrally will be consistent with the test run over a longer time period.

In order to adjust the load state of the test object 2 on the test bench 1 by the load machine 5 according to the test run, the simulation unit 20 sends information to the test bench automation unit 3 and/or the actuator controller. It can be exchanged with (4).

Although the invention has been described based on the example of the exhaust gas system 11, the use of various parts under test PKi of a motor vehicle and others is obviously also possible. Of particular interest is the use of, for example, a combustion engine, a radiator or a power pack of a hybrid vehicle as the component under test (PKi), in each example a plurality of segments (Si) may also be provided. Accordingly, an )), As the test is executed, a trial run is simulated with the car having the test object 2 included in the simulation unit 20. The entire car can be viewed as car parts. The simulation simulates the thermal interaction between the test object (2) and the test object's environment in the form of a heat transfer process, and the test object (2) will experience the course of an actual trial run. However, in particular, any other fictitious heat transfer process can be preset and used equally as a course of test execution. The heat transfer process resulting from this simulation is controlled on the test bench 1 by means of a heat flow actuator 15 _j . The resulting test execution on the test bench (1) is very similar to the real thing.

Claims (8)

In the method for performing a test execution on a test bench (1), a test object (2) in the form of a car or a part of a car is physically installed and operated on the test bench (1) and has a simulation model (22). The simulation unit 20 simulates a test run, wherein at least one temperature is measured at a measurement point MSi as a measured variable MGi of the test object 2 during the test run on the test bench 1 . At least one test target component (PKi) of (2) is divided into a plurality of segments (Si), and during the test, the thermal interaction of the at least one segment (Si) with the environment of the vehicle is determined by the simulation model (22). Simulated by a thermal simulation model 23, wherein the thermal simulation model 23 is a segment heat flow supplied to or lost to at least one segment Si ( ), and calculate the segment heat flow ( ) refers to the heat flow ( A method for performing a test run on a test bench (1), wherein at least one segment (Si) is regulated as a function of the temperature measured by a plurality of heat flow actuators (15 _j ) such that it undergoes 2. The test method of claim 1, wherein the simulation model (22) further comprises one or more of a car model (24), a driver model (25), a road or route model (26), a wheel model (27), and an environment model. Method for performing test runs on a bench (1). Method according to claim 1 or 2, wherein at least one additional measured variable (MGi) of the test object (2) is detected and processed in the simulation model (22). . 3. Test execution according to claim 1 or 2 on a test bench (1), wherein at least one additional measured variable (MGi) of the test object environment of the test object (2) is detected and processed in the simulation model (22). How to do it. In a test bench for performing a test execution, a test object (2) in the form of a car or a part of a car is physically set on the test bench (1), and a simulation unit (20) with a simulation model (22) is used to perform the test. Simulating execution, on a test bench (1), at least one measurement unit (MEi) is provided to a test object (2) which detects the temperature on the test bench (1) as the measured variable (MGi), A simulation model (23) is executed in a simulation unit (20) which simulates, during test execution, the thermal interaction of at least one segment (Si) of the component under test (PKi) of the test object with the environment of the vehicle, The thermal simulation model 23 is a segment heat flow supplied to or lost to at least one segment (Si) ( ) is calculated, and at least one heat flow actuator (15 _j ) is used to determine the heat flow ( ) and is provided on a test bench (1) to undergo _heat flow ( ) by controlling the segment heat flow (Si) in at least one segment (Si) as a function of the measured temperature. A test bench for performing test runs, provided with a heat flow controller 17 that regulates ). 6. Test execution according to claim 5, wherein one or more of the car model (24), driver model (25), road or route model (26), wheel model (27) and environment model are additionally executed in the simulation model (22). A test bench to perform. 7. The test bench (1) according to claim 5 or 6, wherein the test bench (1) is provided with at least one additional measuring unit (MEi) which detects an additionally measured variable (MGi) of the test object (2), said additionally measuring A test bench for performing test execution, where the variable (MGi) is processed by the simulation model (22). 7. The test bench (1) according to claim 5 or 6, wherein at least one additional measurement unit (MEi) is provided on the test bench (1), which detects an additional measured variable (MGi) of the environment of the test object (2). A test bench for performing test execution, where the variable (MGi) measured is processed by the simulation model (22).